Data Center Research Articles

Developing a Third-Party API to Enhance Image Documents at the University of Mosul Data Center

Data integration plays a crucial role in SWE as it guarantees the proper functioning of multiple system, applications, and datasets. This integration is merit since it helps in developing a single reliable source of information from many, and the lack of data integration results in inefficient and time-consuming manual operations such as redundant data input, document retrieval issues, and data handling mistakes. This paper presents a comprehensive methodology for enhancing document images of Mosul University data centers and their quality assessment through integration with a cloud platform. The proposed approach consists of two main stages: the integration stage and the cloud platform stage. During the integration stage, the university's data center is seamlessly connected to a cloud platform via a RESTful API, allowing for efficient data interchange while maintaining data integrity and security. Next, image document representations are received in the cloud platform stage and undergo enhancement processes with specific tools and algorithms designed for image enhancement. The approaches employed in this study encompass image scaling, similarity evaluation utilizing the Histogram of Oriented Gradients (HOG) algorithm, image warping employing the FLANN (Fast Library for Approximate Nearest Neighbors) algorithm, and image quality enhancement by the application of a Laplacian sharpening filter. The proposed integrated cloud-based document enhancement has shown exceptional efficiency in data sharing and precise analysis and enhancement of picture documents within the university's data center.

NVIDIAs Research and Development Investment: Impact on Financial Performance and Market Valuation

This paper provides a detailed financial analysis of NVIDIA Corporation (NVIDIA), a leading technology firm renowned for its advancements in graphics processing units (GPUs), artificial intelligence (AI), data center solutions, autonomous driving, and professional visualization technologies. The analysis delves into NVIDIA's revenue recognition, research and development (R&D) investments, inventory management strategies, and overarching strategic objectives. Utilizing key financial data from fiscal 2024 and the second quarter of fiscal 2025, this study evaluates NVIDIAs recent performance and overall financial health. NVIDIAs substantial investments in AI and data center infrastructure have catalyzed notable revenue growth, underscoring its commitment to pioneering technologies that shape industry standards. Additionally, ongoing R&D efforts have solidified its technological leadership, positioning it as a formidable player in burgeoning markets such as AI, autonomous driving, and the meta-universe. The paper offers a positive outlook on NVIDIAs future growth, highlighting its financial resilience, high profitability, and forward-looking business model. These factors collectively suggest that NVIDIA is well-positioned for sustained long-term success and industry influence.

A Financial Analysis and Valuation of NVIDIA

This paper analyzes Nvidias financial data and its strategic shift towards the AI and data center markets, which the company has recently entered. Nvidia, initially renowned for its GPUs, has now expanded its expertise into AI, computing, and self-driving cars. The calculation of Nvidias financial ratios for the years 2021-2023 reveals strong liquidity, solvency, and profitability indicators, despite external threats such as export restrictions on American microcircuits in China and increased competition. The company has successfully minimized its reliance on debt and is well-positioned to provide AI and cloud services as demand rises. Based on revenue forecasts and free cash flow estimates, the study projects Nvidia's revenue growth rate to be 22% annually, driven by a 30% growth in data centers and AI. However, the analysis also examines other risks, including technological disruptions and market competition. The results indicate that Nvidia is well-positioned to maintain dominance in these rapidly growing markets, but external threats must be addressed to ensure sustained growth.

Establishing bio-logging data collections as dynamic archives of animal life on Earth.

Rapid growth in bio-logging-the use of animal-borne electronic tags to document the movements, behaviour, physiology and environments of wildlife-offers opportunities to mitigate biodiversity threats and expand digital natural history archives. Here we present a vision to achieve such benefits by accounting for the heterogeneity inherent to bio-logging data and the concerns of those who collect and use them. First, we can enable data integration through standard vocabularies, transfer protocols and aggregation protocols, and drive their wide adoption. Second, we need to develop integrated data collections on standardized data platforms that support data preservation through public archiving and strategies that ensure long-term access. We outline pathways to reach these goals, highlighting the need for resources to govern community data standards and guide data mobilization efforts. We propose the launch of a community-led coordinating body and provide recommendations for how stakeholders-including government data centres, museums and those who fund, permit and publish bio-logging work-can support these efforts.

NIST Mass Spectral Libraries in the Context of the Circular Economy of Plastics.

The Mass Spectrometry Data Center (MSDC) has recently started improving existing libraries and creating new ones for identifying and analyzing plastics-related compounds (PRC) and materials (PRM) as part of the NIST circular economy program. PRC are small molecules of dissimilar chemical nature; hence, to increase coverage, we have used three types of ionizations: EI, ESI, and APCI. PRM are solids that include polymers, polymer mixtures, and commercial plastics, so we have used pyrolysis-gas chromatography (py-GC-MS) to create a new searchable library. First, we have increased the coverage of the existing libraries by including as many as possible commercially available PRC. Then, for testing the libraries and to deconvolute complex PRM mixtures, we have analyzed extractable and leachable (E&L) samples and pyrolysis products from one hundred standards of the most common polymers and some of their mixtures using LC-MS/MS, GC-MS, and py-GC-MS. In collaboration with the FDA, the EPA, and other non-government institutions, we are applying techniques, libraries, and tools to areas of interest to the circular economy of plastics, health risk assessments, and environmental challenges.

Comparative Analysis of SARIMA and ETS Models in Forecasting Influenza Cases in China

Accurate prediction of infectious disease cases has always been a research focus in the field of medicine and health. This paper studies the application of time series models in the forecasting of influenza cases in China using the SARIMA and ETS models. The dataset used was collected from the China Public Health Science Data Center from January 2012 to December 2018. The COVID-19 period was excluded to avoid anomalies. Both models were trained on the historical data and tested to see their performance in doing predictions for the year 2018. Based on the evaluation metrics, RMSE and MAPE, the ETS model, one of the best models selected for comparison, was better at modeling the seasonal patterns of influenza, in comparison with SARIMA (0, 1, 0) (1, 0, 0) (12). These findings indicate that the ETS model can make more precise forecasts with wider confidence limits, which may be useful for public health planning and management in China regarding influenza.

A Review of Non-Uniform Load Distribution and Solutions in Data Centers: Micro-Scale Liquid Cooling and Large-Scale Air Cooling

A Review of Non-Uniform Load Distribution and Solutions in Data Centers: Micro-Scale Liquid Cooling and Large-Scale Air Cooling

CLOUD AS INFRASTRUCTURE: THEORISING THE LINKS BETWEEN ‘BIG’ TECH AND ‘SMALL’ TECH

Cloud providers, by creating and owning massive computing assets, produce new internet infrastructure. Large-scale data centres that aggregate hardware resources are an important element shaping the expansion of platform economies. The impact of this configuration of hardware on the dynamics of software development is still unclear. There is growing scholarship on data centres, however the aggregation of a highly scalable hardware system has a profound impact on industry dynamics. Cloud computing lowers the upfront cost of using and owning computing systems. The wide-ranging impact of platforms is explained at least in part by cloud infrastructure: the scalable provisioning of computing resources via the internet. I identify virtualized hardware, modularity, resource sharing, and externalisation as generative of a new internet-based infrastructure. I show these coalesce to form a new industry dynamic, with hardware centralisation on the one hand and decentralisation of cloud users and complement developers on the other. Scholarship in this area is beginning to examine the relations between dominant technology corporations and their networks of users and third-party companies. I contribute to this by examining changing organizational and market dynamics introduced by cloud computing. The findings and analysis are derived from two related studies of changing industrial practices. I focus on the anatomy of the infrastructure and how it activates a new set of relations between start-ups, developers, entrepreneurs, and investors. What is the techno-organisational structure of cloud computing? In answering this, the presentation will offer a view into cloud ecosystems from an industry level and organisational level.

Study on the spatial and temporal evolution of ecosystem service value based on land use change in Xi’an City

Ecosystem service value (ESV) is essential for understanding regional ecological benefits and resources. This study utilizes the fourth phase of land use data from the Resource and Environment Science Data Centre of the Chinese Academy of Sciences. We corrected the ESV coefficient using the equivalent factor method for value per unit area and integrated the biomass factor of farmland ecosystems in Shaanxi Province. This allowed us to adjust the equivalent factor for China’s terrestrial ecosystem on a geographic scale. Based on these corrections, we analyzed changes in land use and the evolution of ecosystem service value over the past two decades in Xi’an, China. Our findings indicate that the proportions of cultivated and forest land in Xi’an remained stable from 2000 to 2020, despite an increase in construction land and a decrease in cultivated areas. Forest and unused lands remained stable, while grassland and water bodies fluctuated. The ESV related to land use in Xi’an increased by 938.8 million yuan during this period, with high-value areas primarily located in the forested regions south of the Qinling Mountains and along the Weihe, Bahe, and Chanhe Rivers. Low–value zones were concentrated in the urban core. This research enhances methodologies for quantifying urban ESV, providing vital support for land resource management, ecological conservation, and high-quality urban development in major cities in China. These findings will inform policy-making for sustainable urban growth.

Persistence of advanced therapies in patients with inflammatory bowel disease: retrospective cohort study using a large healthcare claims database in Japan.

There are few studies that comprehensively report real-world persistence for first-line advanced therapies used to treat inflammatory bowel disease. We aimed to describe persistence of first-line advanced therapies among incident biologic or Janus kinase inhibitor users with inflammatory bowel disease. Retrospective cohort study using the Japan Medical Data Center database from January 1, 2010, until September 30, 2022. Patients aged ≥15 years with relevant diagnostic and treatment codes were included. All eligible patients were observed until study end (September 30, 2022), death, or disenrollment, whichever occurred first. Among 1,115 patients with Crohn's disease included in the analysis, 41.4% initiated adalimumab, 37.4% infliximab, 18.1% ustekinumab, and 3.0% vedolizumab. Median age was 31.2-34.8 years, 72.8% to 85.9% were male. Persistence at 12 months was 84.7% for adalimumab, 87.7% for infliximab, 91.3% for ustekinumab, and 53.1% for vedolizumab. Persistence at 24 months was 76.3%, 76.8%, 80.4%, and 28.6%, respectively. Among 1,942 patients with ulcerative colitis, 24.8% initiated adalimumab, 33.6% infliximab, 11.2% golimumab, 17.5% vedolizumab, 5.6% ustekinumab, and 7.3% tofacitinib. Mean age was 38.2-40.4 years, 57.4% to 65.8% were male. Persistence at 12 months was 57.6% for adalimumab, 87.7% for infliximab, 54.9% for golimumab, 69.7% for vedolizumab, and 84.0% for ustekinumab. At month 24, persistence for ustekinumab was 75.0%, versus 42.9%-59.4% for other treatments. Index treatment with ustekinumab resulted in high persistence through 24 months after initiation in patients with Crohn's disease or ulcerative colitis. Our study provides insights into the real-world usage of advanced treatments for patients with IBD in Japan.

Toward climate neutral data centers: Greenhouse gas inventory, scenarios, and strategies.

The climate impact of data centers is expected to increase due to rising demand for information and communication technology services. At the same time, the European Union aims for climate neutral data centers by 2030. To map potential developments of emissions associated with data centers to the year 2030, we develop a generic data center greenhouse gas (GHG) inventory in accordance with the GHG protocol. Our results show that climate neutrality cannot be achieved by reduction measures only. We predict an increase in energy use for our illustrative Germany-based data center by 20%, and an increase in the overall carbon footprint by 13% between 2020 and 2030. Although operational measures and increased use of renewable electricity can decrease GHG emissions over the 2020 level by up to 70%, achieving net zero inevitably requires mechanisms to mitigate and compensate emissions across the value chain.

Investigating the Environmental Sustainability of Data Centers

Data centers play a critical role in supporting global IT infrastructure, representing a rapidly growing industry. However, this growth is coupled with significant environmental consequences, such as the consumption of 1-2% of the annual global energy supply, increased carbon emissions, and the loss of valuable agricultural land. This study reviews solutions that address these impacts by evaluating sustainable practices in data center construction, design, and maintenance. The results point to several key strategies that improve environmental sustainability in these categories. In construction, replacing traditional materials, such as concrete and steel, with eco-friendly alternatives such as green concrete and steel slag can reduce carbon footprints. Additionally, permeable pavements can manage the stormwater runoff that was previously absorbed by the farmland on which data centers are built. Design improvements, including underfloor air distribution, economizers, ice storage, and energy recovery wheels can optimize ventilation systems, lowering energy demands by up to 25%. Sustainable maintenance practices emphasize the use of advanced liquid cooling systems, as well as thermal energy storage to reduce peak electricity usage and incorporate more renewable energy sources. Overall, it is vital to implement strategies that prioritize sustainability in data centers in relation to the construction, design, and maintenance to reduce the environmental footprint of this growing industry.

Performance of a multi-cooling sources cooling system with photovoltaics and waste heat recovery in data center

Performance of a multi-cooling sources cooling system with photovoltaics and waste heat recovery in data center

Influence of Indonesian Army Information System Arrangement Concept in Realizing One Integrated, Fast, and Secure Data

Bureaucratic Reform (RB) based on Presidential Decree Number 81 of 2010 concerning the Grand Design of Bureaucratic Reform 2010-2025 in practice must be carried out by all Ministries/Institutions/Dinas/Institutions (K/L/D/I) in an innovative, creative and monitored manner institutional. The government also then issued Presidential Regulation Number 39 of 2019 concerning Data Management. Indonesia's One Data Policy (SDI) is expected to be a strategy for improving data governance to produce data that is accurate, up-to-date, integrated and accountable so that it can become the foundation for determining effective and appropriate policies. Indonesia's One Data Policy (SDI) is expected to be a strategy for improving management data to produce accurate, timely, integrated and accountable data which forms the basis for the creation of effective and targeted policies. The TNI AD Data Center is an Indonesian military facility that contains electronic systems and related components to be used and used for the placement, storage and processing of data in support of the Indonesian military information system. Data center facility infrastructure includes four aspects. First, infrastructure that guarantees the continuity of business processes when something critical and dangerous occurs in the data center. Internalization of RB, SPBE, and realizing SDI within the TNI AD institution must be realized through the arrangement of the Information System in order to further support the implementation of the Main Tasks of the TNI AD. The TNI AD needs to conceptually manage the data center, data processing architecture, network centralization, human resources, software and organizational management.

Comparing Resources Distribution Techniques in Cloud Computing to Improve Quality of Service

The Resources Distribution Approach is considered a major issue for enhancing quality of service (QoS) in data centers. It helps in achieving the goals for resources cost by getting a proper balance of conflicting requests. Resources distribution should also be performed among various data centers to get the shortest acceptable time required to honor requests from clients. In this paper, we compare distribution resources techniques that aim to enhance QoS parameters like the average reply time (ART), average server processing time (ASPT) and cost of virtual machine (VM). We use three of resources distribution techniques: Closest Data Center (CDC), Optimal Replay Time (ORT), and Reconfigure Dynamically (RD). We study the activity of resource distribution techniques based on acceptable metrics and examined their drawbacks. Our main goal is to analyze broker policies that help cloud customers to improve QoS and reduce the load of requests for the tasks in the network.

Analisis Pembatasan Kewenangan Bawaslu dalam Sentra Gakkumdu Perspektif Sosiologi Fungsionalis

The enforcement of election crimes committed by Bawaslu is the reality that occurs that not always elections can run directly, freely, confidentially, honestly, and fairly. Based on the analysis of conditions that are mainly studied through the reality that occurs, the obstacles that often occur are differences in perception between members of the Gakkumdu Center, loss of evidence of conjecture, limited time, and limited Bawaslu human resources. Through Law No. 22 of 2007, the existence of Bawaslu is no longer subordinate to the KPU. The above conditions then initiated the birth of development research in the framework of claiming novelty, parsing the role of each institution in the Gakkumdu Center through a new point of view, namely functionalist sociology. The function of Sentra Gakkumdu is as a coordination forum in the process of handling every violation of election crimes, implementation of the election criminal pattern itself, data centers, competency improvement, evaluation monitoring. Therefore, it is necessary to coordinate and collaborate to build between law enforcement members in the Gakkumdu Center.

Stochastic optimization of combined energy and computation task scheduling strategies of hybrid system with multi-energy storage system and data center

Stochastic optimization of combined energy and computation task scheduling strategies of hybrid system with multi-energy storage system and data center

Enhancing intrusion detection against denial of service and distributed denial of service attacks: Leveraging extended Berkeley packet filter and machine learning algorithms

AbstractAs organizations increasingly rely on network services, the prevalence and severity of Denial of Service (DoS) and Distributed Denial of Service (DDoS) attacks have emerged as significant threats. The cornerstone of effectively addressing these challenges lies in the timely and precise detection capabilities offered by advanced intrusion detection systems (IDS). Hence, an innovative IDS framework is introduced that seamlessly integrates the extended Berkeley Packet Filter (eBPF) with powerful machine learning algorithms—specifically Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), and TwinSVM—enabling unparalleled real‐time detection of DDoS attacks. This cutting‐edge solution provides a robust and scalable IDS framework to combat DoS and DDoS threats with high efficiency, leveraging eBPF's capabilities within the Linux kernel to bypass typical user space constraints. The methodology encompasses several key steps: (a) Collection of data from the renowned CIC‐IDS‐2017 repository; (b) Processing the raw data through a meticulous series of steps, including transmission, cleaning, reduction, and discretization; (c) Utilizing an ANOVA F‐test for the extraction of critical features from the preprocessed data; (d) Application of various ML algorithms (DT, RF, SVM, and TwinSVM) to analyze the extracted features for potential intrusion; (e) Implementing an eBPF program to capture network traffic and harness trained model parameters for efficient attack detection directly within the kernel. The experimental results reveal outstanding accuracy rates of 99.38%, 99.44%, 88.73%, and 93.82% for DT, RF, SVM, and TwinSVM, respectively, alongside remarkable precision values of 99.71%, 99.65%, 84.31%, and 98.49%. This high‐speed, accurate detection model is ideally suited for high‐traffic environments such as data centers. Furthermore, its foundational architecture paves the way for future advancements, including the potential integration of eBPF with XDP to achieve even lower‐latency packet processing. The experimental code is available at the GitHub repository link: https://github.com/NemalikantiAnand/Project.

Advanced Security Measures for Virtualized Data Storage: A Comprehensive Study

With the widespread application of virtualization technology in data centers, the security issues of data storage have become increasingly prominent. This paper aims to explore the application of information security in virtualized data storage to address the increasingly complex security challenges. The article first reviews the development of virtualization technology and the key technologies of virtualized data storage, then analyzes the current state and threats faced by information security in virtualized environments. Based on this, the paper discusses security requirements such as data integrity, access control, and privacy protection in detail, and delves into the application of information security technologies such as data encryption, intrusion detection and defense systems, security information and event management, and virtualization security policy management. Furthermore, this paper proposes a comprehensive security framework for virtualized data storage and describes its design principles, component architecture, and implementation deployment strategies in detail. Through experimental design and case studies, the paper validates the effectiveness of the proposed framework and evaluates its performance. Finally, the paper summarizes the research findings and proposes prospects for future research directions. This study not only provides theoretical support and practical guidance for information security in virtualized data storage but also offers valuable references for researchers and practitioners in related fields.

Innovative Retrofits for enhanced thermal performance in data centers using Independent Row-Based cooling systems

Innovative Retrofits for enhanced thermal performance in data centers using Independent Row-Based cooling systems